Working memory refers to the active mental retention of information, to its manipulation, and to its use in guiding behavior. Its role in many types of high level cognition, as a factor underlying individual differences across a broad spectrum of experimental and real world measures, and as a factor in many psychiatric and neurological diseases, makes it important to many branches of cognitive and clinical neuroscience. The motivation for this Revision Application comes from progress that the Principal Investigator's group has made in addressing Specific Aim 1 from the parent grant, R01 MH095984. That Aim is To test the hypothesis that delay-period oscillatory activity carries stimulus-specific information, and although the results of three studies do support this broad hypothesis, they do so with an important qualification: Active neural representation of a stimulus may result from its attentional selection, rather than from its storage, per se, in working memory. More specifically, multivariate pattern analyses (MVPA) of data from three types of experiment -- functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and combined EEG and transcranial magnetic stimulation (TMS) - have failed to find evidence for the active retention of information that is in working memory, but outside the focus of attention. These are intriguing results, which suggest that the brain may use a passive, structural trace to hold information in working memory. If true, this fact would require a reconsideration of over 50 years of dogma about how working memory works. However, caution is warranted, because these recent findings are subject to two important qualifications: First, they comprise a set of null results; Second, each was acquired with a noninvasive method that is inherently limited in spatial and temporal resolution. The motivation for the new experiment proposed in this Revision Application, therefore, is to overcome the inherent limitations of the studies summarized above by employing a physiological method, electrocorticographic (ECoG) recording from neurosurgical patients, that features spatial and temporal resolution that are far superior to those of fMRI or EEG. Specifically, adoption of ECoG for this new study will enable a test of the hypothesis that a mechanism of phase-encoded selection among multiple active representations may underlie the attentional prioritization of one among many items held in working memory. A similar mechanism has recently been discovered, from ECoG data recorded during performance of a hierarchical cognitive control task, by an investigator who will serve as a Co-Investigator on this experiment. This newly proposed experiment fits with the Aims of the parent project and thus its findings may also have translational relevance for the diagnosis and treatment of psychiatric disorders that are associated with abnormal neural oscillations.